Skip to main content

Potential prognostic value of PD-L1 and NKG2A expression in Indonesian patients with skin nodular melanoma



Biomarker mRNA levels have been suggested to be predictors of patient survival and therapy response in melanoma cases. This study aimed to investigate the correlations between the mRNA expression levels of PD-L1 and NKG2A in melanoma tissue with clinicopathologic characteristics and survival in Indonesian primary nodular melanoma patients.


Thirty-one tissue samples were obtained; two were excluded from survival analysis due to Breslow depth of less than 4 mm. The median survival of upregulated and normoregulated PD-L1-patients were 15.800 ± 2.345 and 28.945 ± 4.126 months, respectively. However, this difference was not significant statistically (p = 0.086). Upregulated and normoregulated NKG2A patients differed very little in median survival time (25.943 ± 7.415 vs 26.470 ± 3.854 months; p = 0.981). Expression of PD-L1 and NKG2A were strongly correlated (rs: 0.787, p < 0.001). No clinicopathologic associations with PD-L1 and NKG2A mRNA levels were observed. These results suggest that PD-L1 may have potential as a prognostic factor. Although an unlikely prognostic factor, NKG2A may become an adjunct target for therapy. The strong correlation between PD-L1 and NKG2A suggests that anti-PD-1 and anti-NKG2A agents could be effective in patients with PD-L1 upregulation. The mRNA levels of these two genes may help direct choice of immunotherapy and predict patient outcomes.


Among cutaneous malignancies, melanoma results in the highest mortality; statistics revealed that melanomas caused 287,723 new cases and 60,712 deaths worldwide in 2018 [1]. These numbers are expected to increase, with an estimated 340,721 new cases projected in 2025 [2]. Trials of various systemic chemotherapy combinations did not improve survival significantly. Thus, current research is focused on new therapeutic agents, including immune checkpoint blockers.

Agents that block immune checkpoints, such as PD-1/PD-L1, help improve the immune response to cancerous cells. Nivolumab, an anti-PD-1 antibody, has been approved by the FDA for the treatment of advanced melanomas [3]. Other immunotherapies, such as monalizumab, a humanized anti-NKG2A antibody that enhances NK cell and CD8+ T cell activity [4], are also under development. These drugs show great promise and more durable responses compared with targeted therapy agents.

Before starting anti-PD-1 immunotherapy, clinicians commonly test tumor tissues for PD-L1 expression by using immunohistochemistry (IHC). Tumors expressing PD-L1 respond better to anti-PD-1/PD-L1 agents compared with non-expressers [5]. However, the results of recent studies on the effect of PD-L1 expression on survival are conflicting [6]. Detection using IHC also presents several limitations, such as the varying performance of different antibodies, nonstandard cut-off values, and operator dependence [7, 8].

Researchers are exploring new methods to predict therapy responses and survival in melanoma patients. Several studies have investigated the use of biomarker mRNA levels as an alternative parameter [9,10,11]. Gupta et al. reported that mRNA levels of PD-L1/2 show potential in predicting survival and response toward immunotherapy in metastatic melanoma [12]. Given the emergence of monalizumab, the potential prognostic and therapeutic roles of NKG2A should also be investigated.

Immuno-oncological research is still rare in Indonesia. Most Indonesian patients are treated with surgical resection, dacarbazine chemotherapy, and radiotherapy. Information on the expression of immune-checkpoint molecules is needed to gauge the potential efficacy of using immunotherapy agents in Indonesia. Thus, this study aimed to investigate the prognostic role of mRNA levels of PD-L1 and NKG2A, as well as the associated clinicopathologic characteristics.

Main text

Materials and methods

Formalin-fixed paraffin-embedded (FFPE) tissue samples from patients diagnosed in 2012–2019 with primary cutaneous nodular melanoma were collected from the archives of the Department of Anatomical Pathology, Dr. Sardjito Hospital, which is the main cancer referral center in Yogyakarta, Indonesia. Cases with prior chemotherapy or radiotherapy, incomplete clinical data, and degraded specimens were excluded. Thirty-one samples were analyzed in this retrospective cohort study, and all patients were of Javanese ethnicity.

RNA was extracted from FFPE tissues using GeneAll® Ribospin™ II (GeneAll Biotechnology, Seoul, South Korea). Real-time polymerase chain reaction (RT-PCR) for PD-L1 and NKG2A expression quantification was conducted using AccuPower® GreenStar™ RT-qPCR PreMix on an Exicycler™ 96 (Bioneer Corp., Daejeon, South Korea) with primer pairs and thermocycler conditions as previously described by Vassilakopoulou et al. and Meckawy et al. [13, 14]. The expressions of PD-L1 and NKG2A were calculated from the quantification cycle (Cq) values of the gene targets and normalized against GAPDH as an internal control. Subsequent normalization was performed using the ΔΔCq values of RNA derived from healthy skin tissues. Age, sex, tumor location, Breslow thickness, greatest diameter, lymph node involvement, and stage were retrieved from medical records. Pathological data, including the presence of necrosis, lymphovascular invasion, tumor-infiltrating lymphocytes (TILs), and mitotic index, were obtained from hematoxylin–eosin and Ki-67 IHC stained slides. Survival status (living or deceased) was determined through telephone calls at the point of follow-up of the study (until April 2020).

Samples were classified as normoregulated if the expression was lower than or equal to the mean of the PD-L1 and NKG2A levels; conversely, samples were classified as upregulated if the expression was above the mean. Comparison of mRNA level averages based on categorical clinicopathologic characteristics was performed using Mann–Whitney U tests. Spearman correlation was used to analyze associations between the expression of PD-L1 and NKG2A and continuous clinicopathologic features. Kaplan–Meier analysis and log-rank tests with Cox regression were used to determine hazard ratios (HRs) for survival analysis. To minimize the effect from Breslow thickness, two samples was excluded from survival analysis due to depth of less than 4 mm.


The characteristics of the subjects are presented in Table 1. Most tumors were located on the extremities (70.97%) and thicker than 4 mm (93.55%). Necrosis and TILs were present in 74.19% of the samples, respectively. The clinical stages were evenly distributed among stages II (29.03%), III (35.48%), and IV (35.48%).

Table 1 Clinicopathologic characteristics of the subjects

The expression of PD-L1 and NKG2A was not significantly associated with the patients’ clinicopathologic characteristics (Additional file 1: Table S1). Spearman correlation showed that NKG2A and PD-L1 mRNA levels were strongly correlated (Additional file 2: Table S2).

In the Cox univariate regression analysis, higher stage, upregulated PD-L1, and upregulated NKG2A were related to higher risks of death, with respective HR of 1.080 (p = 0.763), 2.429 (p = 0.101), and 1.011 (p = 0.981) (Additional file 3: Table S3). In multivariate analysis, the HR for PD-L1 increased to 3.488 (p = 0.066), while the other HRs decreased to 1.017 (p = 0.951) for stage and 0.590 (p = 0.391) for NKG2A upregulation. However, the differences were not statistically significant.

Patients with normoregulated PD-L1 expression had longer median survival time (28.945 ± 4.126 months) compared with upregulated expressers (15.800 ± 2.345 months; p = 0.086) (Fig. 1). Similar findings were observed for the normoregulated (26.470 ± 3.854 months) and upregulated NKG2A subjects (25.943 ± 7.415 months; p = 0.981) (Fig. 2). However, both differences were not significant statistically. Presence of TILs did not affect the survival curves significantly (p = 0.422) (Additional file 4: Fig. S1). The survival curves of the upregulated and normoregulated groups for PD-L1 and NKG2A did not differ significantly when divided based on the presence of TILs (Additional file 5: Fig. S2 and Additional file 6: Fig. S3).

Fig. 1
figure 1

Kaplan–Meier survival curves comparing the survival of patients with primary nodular melanoma with upregulated or normoregulated PD-L1 expression (p = 0.086)

Fig. 2
figure 2

Kaplan–Meier survival curves comparing the survival of patients with primary nodular melanoma with upregulated or normoregulated NKG2A expression (p = 0.981)


In this study, we compared the clinicopathologic characteristics and overall survival of Indonesian primary nodular melanoma cases with different mRNA levels of PD-L1 and NKG2A. Two important findings were observed: (1) cases with PD-L1 overexpression tended to have lower survival rates and (2) PD-L1 and NKG2A levels were strongly correlated.

Melanoma patients with upregulated PD-L1 tended to have lower overall survival, with an approximately two to threefold higher HR compared to normal expressers. This trend was not proven statistically significant, possibly due to the small sample size. However, the increase in the HR for PD-L1 upregulation in multivariate regression suggests this trend may exist in the overall population of melanoma patients in Indonesia.

The PD-L1 molecule interacts with PD-1 receptors on T cells, causing anergy, exhaustion, and apoptosis [15]. Melanoma cells can thus evade the immune system by increasing their PD-L1 expression. Melanoma cell line studies show that cells with upregulated PD-L1 demonstrate highly invasive and aggressive behavior [16]. In a study on melanoma patients treated with surgery and dacarbazine, patients with positive PD-L1 on IHC staining had lower median survival time compared with the subgroup with negative or indeterminate PD-L1 status (9.7 months vs. 11.6 months) [17].

Our results seem to oppose those of Gupta et al., who observed that higher PD-L1 mRNA levels reflect better prognoses for melanoma patients treated with anti-PD-1 agents [12]. This disparity may have stemmed from differences in the treatment regimen. Patients with high expression of PD-L1 respond well to anti-PD-1 antibodies, hence the increase in survival [5]. The results suggest that PD-L1 expression is a negative prognostic factor in patients with melanoma treated with conventional chemotherapy. However, when treated with anti-PD-1 antibodies, patients with high levels of PD-L1 respond well and have good outcomes. Therefore, the choice of therapy also affects the performance of PD-L1 as a prognostic factor.

Tumor cells can express PD-L1 independently (called constitutive expression) or in response to TILs (reactive expression). TILs can secrete interferon-gamma, which induces the expression of PD-L1 in tumor cells [15]. When PD-L1 expression is accompanied by the presence of TILs, the expression is likely reactive, and vice versa. Due to the strongly positive correlation between PD-L1 and TILs, studies report that most melanomas express PD-L1 reactively [18]. However, in our study, the proportion of upregulated PD-L1 among the cases without TILs was higher (37.5%) than the cases with TILs (25%). The groups without TILs also had higher average PD-L1 mRNA levels than the group with TILs. This finding suggests that the cases with upregulated PD-L1 in our study are likely constitutive expressers.

These two modes of PD-L1 expression may have different prognostic implications. When previous studies divided patients based on PD-L1 expression and the presence of TILs, patients with constitutive PD-L1 expression without lymphocyte infiltrates showed the poorest outcomes, followed by reactive PD-L1 expressers, those with PD-L1(−) without TILs, and, finally, those with PD-L1(−) and TILs [16, 19]. In our results, both groups with upregulated PD-L1 showed poorer prognoses than the groups with normoregulated PD-L1. However, survival did not differ significantly when the cases were divided further based on TIL status, likely because none of the patients were treated using immunotherapy, in which the presence of TILs predicts improved response [20]. Another factor that could explain this lack of significance is the limited sample size.

Upregulated NKG2A mRNA did not appear to affect survival, given the highly insignificant results in univariate and multivariate regression analysis. NKG2A is an inhibitory receptor found on NK cells [21]. Cancer cells can attempt to evade the immune system by upregulating HLA self-molecules that activate NKG2A receptors and impair the function of NK cells. Trials in mouse models indicate that monalizumab is ineffective as a single therapy but highly effective when used together with other immunotherapy agents that promote activated TILs, such as anti-PD-1 or cancer vaccines [22]. One escape strategy used by cells to escape cytotoxic TILs is downregulation of MHC I expression, which renders them targets for NK cells [23]. This finding may explain the role of anti-NKG2A as an adjunct treatment for other immunotherapies. Our results reinforce the idea that NKG2A may not be an independent therapeutic target and prognostic factor but may play a role in combination therapy.

The mRNA expressions of NKG2A and PD-L1 were strongly correlated. This finding indicates that tumors with high PD-L1 expression would also likely express NKG2A strongly and, thus, respond well to anti-NKG2A agents. When NKG2A expression was combined with the TIL parameter, the distribution of survival curves obtained resembled the curves for PD-L1 combined with TILs. NKG2A upregulation with and without the presence of TILs may have different pathogeneses and prognostic implications, like PD-L1.

The lack of correlation between the expression of NKG2A and PD-L1 and clinicopathologic characteristics in this work resembles the findings of several previous studies [6, 18]. The small sample size of this study may have contributed to the low statistical significance found.

The findings of this study must be interpreted with caution especially due to the small sample size. However, our results support the findings of several studies that show that mRNA profiles may serve as a prognostic factor in melanoma cases [12, 24]. Further research and clinical trials are needed to ascertain the roles of PD-L1 and NKG2A in the prognosis and therapy of Asian patients who have not previously received immune checkpoint inhibitors.


We investigated the correlations between the mRNA levels of PD-L1 and NKG2A with clinicopathologic characteristics and survival in primary nodular melanoma patients in Yogyakarta, Indonesia. Patients with upregulated PD-L1 expression had shorter median overall survival, although insignificant statistically. PD-L1 and NKG2A mRNA levels were positively correlated.

Our findings suggest that the therapy regimen and presence of TILs may affect the prognostic role of PD-L1 expression. NKG2A was not proven to be an independent predictive factor but may serve as an adjunct target for therapy. The strong correlation between PD-L1 and NKG2A suggests that anti-PD1 and anti-NKG2A agents may be effective in patients with PD-L1 upregulation. Studies with larger subject groups are needed to confirm the patterns of PD-L1 expression in Asian cases.


Our study was limited by its small sample size and homogenous ethnic population. Results among diverse Indonesian and Asian populations may differ. TILs examination did not discriminate between lymphocyte subtypes.

Availability of data and materials

This submission contains all of the data analyzed during the study. Unprocessed data can be requested from the corresponding author.



Cluster of differentiation


Quantification cycle


Deoxyribonucleic acid


Formalin-fixed paraffin-embedded


Glyceraldehyde 3-phosphate dehydrogenase


Human leukocyte antigen




Major histocompatibility complex


Messenger ribonucleic acid


Natural killer


Natural killer group 2A


Programmed death-1


Programmed death-ligand 1


Real-time polymerase chain reaction


Tumor-infiltrating lymphocytes


  1. International Agency for Reserch on Cancer. GLOBOCAN fact sheet: melanoma of skin. Global Cancer Observatory. 2018.

  2. International Agency for Research on Cancer. Estimated number of incident cases from 2018 to 2040: Melanoma of skin. Cancer Tomorrow. 2018.

  3. Domingues B, Lopes JM, Soares P, Pópulo H. Melanoma treatment in review. ImmunoTargets Ther. 2018;7:35–49.

    Article  CAS  Google Scholar 

  4. Van Hall T, André P, Horowitz A, Ruan DF, Borst L, Zerbib R, et al. Monalizumab: inhibiting the novel immune checkpoint NKG2A. J Immunother Cancer. 2019;7(1):263.

    Article  Google Scholar 

  5. Abdel-Rahman OPD. PD-L1 expression and outcome of advanced melanoma patients treated with anti-PD-1/PD-L1 agents: a meta-analysis. Immunotherapy. 2016;8(9):1081–9.

    Article  CAS  Google Scholar 

  6. Yang J, Dong M, Shui Y, Zhang Y, Zhang Z, Mi Y, et al. A pooled analysis of the prognostic value of PD-L1 in melanoma: evidence from 1062 patients. Cancer Cell Int. 2020;20(1):96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Patel SP, Kurzrock R. PD-L1 expression as a predictive biomarker in cancer immunotherapy. Mol Cancer Ther. 2015;14(4):847–56.

    Article  CAS  Google Scholar 

  8. Lee HH, Wang YN, Xia W, Chen CH, Rau KM, Ye L, et al. Removal of N-linked glycosylation enhances PD-L1 detection and predicts anti-PD-1/PD-L1 therapeutic efficacy. Cancer Cell Int. 2019;36(2):168-178.e4.

    Article  CAS  Google Scholar 

  9. Tang Y, Xie C, Zhang Y, Qin Y, Zhang W. Overexpression of mRNA-decapping enzyme 1a predicts disease-specific survival in malignant melanoma. Melanoma Res. 2018;28(1):30–6.

    Article  CAS  Google Scholar 

  10. Hoffmann F, Zarbl R, Niebel D, Sirokay J, Fröhlich A, Posch C, et al. Prognostic and predictive value of PD-L2 DNA methylation and mRNA expression in melanoma. Clin Epigenet. 2020;12(1):94.

    Article  CAS  Google Scholar 

  11. Wan X, Liu R, Li Z. The prognostic value of HRAS mRNA expression in cutaneous melanoma. BioMed Res Int. 2017;2017:5356737.

    PubMed  PubMed Central  Google Scholar 

  12. Gupta S, McCann L, Chan YGY, Lai EW, Wei W, Wong PF, et al. Closed system RT-qPCR as a potential companion diagnostic test for immunotherapy outcome in metastatic melanoma. J Immunother Cancer. 2019;7(1):254.

    Article  Google Scholar 

  13. Vassilakopoulou M, Avgeris M, Velcheti V, Kotoula V, Rampias T, Chatzopoulos K, et al. Evaluation of PD-L1 expression and associated tumor-infiltrating lymphocytes in laryngeal squamous cell carcinoma. Clin Cancer Res. 2016;22(3):704–13.

    Article  CAS  Google Scholar 

  14. Meckawy GR, Mohamed AM, Zaki WK, Khattab MA, Amin MM, ElDeeb MA, et al. Natural killer NKG2A and NKG2D in patients with colorectal cancer. J Gastrointest Oncol. 2019;10(2):218–25.

    Article  Google Scholar 

  15. Taube JM, Anders RA, Young GD, Xu H, Sharma R, McMiller TL, et al. Colocalization of inflammatory response with B7–H1 expression in human melanocytic lesions supports an adaptive resistance mechanism of immune escape. Sci Transl Med. 2012;4(127):127ra37.

    Article  Google Scholar 

  16. Audrito V, Serra S, Stingi A, Orso F, Gaudino F, Bologna C, et al. PD-L1 up-regulation in melanoma increases disease aggressiveness and is mediated through miR-17-5p. Oncotarget. 2017;8(9):15894–911.

    Article  Google Scholar 

  17. Robert C, Long GV, Brady B, Dutriaux C, Di Giacomo AM, Mortier L, et al. Five-year outcomes with nivolumab in patients with wild-type BRAF advanced melanoma. J Clin Oncol. 2020;38(33):3937–46.

    Article  CAS  Google Scholar 

  18. Obeid JM, Erdag G, Smolkin ME, Deacon DH, Patterson JW, Chen L, et al. PD-L1, PD-L2 and PD-1 expression in metastatic melanoma: correlation with tumor-infiltrating immune cells and clinical outcome. Oncoimmunology. 2016;5(11):e1235107.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Massi D, Brusa D, Merelli B, Falcone C, Xue G, Carobbio A, et al. The status of PD-L1 and tumor-infiltrating immune cells predict resistance and poor prognosis in BRAFi-treated melanoma patients harboring mutant BRAFV600. Ann Oncol. 2015;26(9):1980–7.

    Article  CAS  PubMed  Google Scholar 

  20. Badalamenti G, Fanale D, Incorvaia L, Barraco N, Listì A, Maragliano R, et al. Role of tumor-infiltrating lymphocytes in patients with solid tumors: can a drop dig a stone? Cell Immunol. 2017;2019(343):103753.

    Article  CAS  Google Scholar 

  21. Zaghi E, Calvi M, Marcenaro E, Mavilio D, Di Vito C. Targeting NKG2A to elucidate natural killer cell ontogenesis and to develop novel immune-therapeutic strategies in cancer therapy. J Leukoc Biol. 2019;105(6):1243–51.

    Article  CAS  Google Scholar 

  22. Borst L, van der Burg SH, van Hall T. The NKG2A–HLA-E axis as a novel checkpoint in the tumor microenvironment. Clin Cancer Res. 2020;26(21):5549–56.

    Article  CAS  Google Scholar 

  23. Lee H, Quek C, Silva I, Tasker A, Batten M, Rizos H, et al. Integrated molecular and immunophenotypic analysis of NK cells in anti-PD-1 treated metastatic melanoma patients. Oncoimmunology. 2019;8(2):e1537581.

    Article  PubMed  Google Scholar 

  24. Jayawardana K, Schramm SJ, Haydu L, Thompson JF, Scolyer RA, Mann GJ, et al. Determination of prognosis in metastatic melanoma through integration of clinico-pathologic, mutation, mRNA, microRNA, and protein information. Int J Cancer. 2015;136(4):863–74.

    Article  CAS  Google Scholar 

Download references


We would like to thank Nur Eka Wiraditya (Department of Anatomical Pathology, Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada, Yogyakarta, Indonesia) for assisting us with the laboratory procedures in this study.


This study was funded by Universitas Gadjah Mada through Program Rekognisi Tugas Akhir 2020 (2607/UN1/DITLIT/DIT-LIT/PT/2020). Laboratory reagents and consumables constituted most of the expenditures. Publication costs were covered by the Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada.

Author information

Authors and Affiliations



RDS designed the study and performed the analysis. HTR, MFP, SLA, KS, and TA wrote the manuscript. YI and MRR contributed to the data collection. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Hanggoro Tri Rinonce.

Ethics declarations

Ethics approval and consent to participate

This study complied with the Declaration of Helsinki and the Belmont Report. The protocol for this study was approved by the Medical and Health Research Ethics Committee of the Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada (KE/FK/0599/EC/2020). The study was conducted with formal permission from the appropriate hospital officials after explaining the research objectives and procedures. Patients were informed preoperatively of the use of data and tissue samples for research. All patients have consented in written forms.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

Comparison of PD-L1 and NKG2A expression levels based on clinicopathologic characteristics and overall survival.

Additional file 2: Table S2.

Spearman correlations between continuous variables.

Additional file 3: Table S3.

Univariate and multivariate Cox regression results for stage, PD-L1 upregulation, and NKG2A upregulation.

Additional file 4: Fig. S1.

Kaplan–Meier survival curves comparing the survival of patients with primary nodular melanoma with or without TILs.

Additional file 5: Fig. S2.

Kaplan–Meier survival curves comparing the survival of patients with primary nodular melanoma based on PD-L1 expression and the presence of TILs.

Additional file 6: Fig. S3.

Kaplan–Meier survival curves comparing the survival of patients with primary nodular melanoma based on NKG2A expression and the presence of TILs.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Saputro, R.D., Rinonce, H.T., Iramawasita, Y. et al. Potential prognostic value of PD-L1 and NKG2A expression in Indonesian patients with skin nodular melanoma. BMC Res Notes 14, 206 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: